Download Ovarian Cancer Screening - University of Kentucky | Medical Center

Survey
yes no Was this document useful for you?
   Thank you for your participation!

* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project

Document related concepts
no text concepts found
Transcript
Ovarian Cancer Screening
J. R. van Nagell, Jr., M.D.
Professor and Director
Division of Gynecologic Oncology
University of Kentucky Ovarian Cancer
Screening Research Program
Ovarian Cancer






2006: 20,180 new cases/yr; 15,310 deaths
Fifth leading cause of cancer death among women
Leading cause of death among gynecologic
malignancies
5 yr survival: Stage I - ~ 90%
Stage III/IV - 20%
Prevalence- 50/100,00 in women > 50 yrs.
75% cases diagnosed with advanced disease
Ovarian Cancer Symptoms






Abdominal bloating
Abdominal pain
Indigestion
Urinary frequency
22% of patients ignored symptoms
Incorrect diagnosis 30%
Goff BA et al. Cancer 89:2068-2075, 2000.
Ovarian Cancer
Target Symptoms
Abdominal pain (30%) – OR 6.0
 Abdominal swelling (16.5%) – OR 30
 Gastrointestinal symptoms (8.5%) – OR 2.3
 Pelvic pain (5.4%) – OR 4.3

Smith LH et al. Cancer 104:1398-1407, 2005.
Delay in Diagnosis



On average patients seek medical attention
9 months after onset of symptoms
On average patients receive a pelvic exam
9 months after seeking medical attention
Gilda Radner, Ella Grasso, Madeline Kahn, Liz
Tilberis, Cassandra Hanis-Brosnan, Coretta Scott
King, Patsy Ramsey, Loretta Young, Dinah Shore,
Jessica Tandy, Lauro Nyro, Joan Hackett, Dixie
Lee, Rosalind Franklin (discoverer of DNA),
Sandy Dennis; Bess Myerson & Carol Channing
are survivors
Sackett, et. al., Clinical Epid., Boston: Little, Brown & Co., 1985
http://en.wikipedia.org/wiki/Ovarian_cancer#Notable_victims_of_ovarian_cancer
Accuracy of Pelvic Examination




289 ovaries evaluated clinically in 151 women
under anesthesia
Ovaries detected clinically in 30% of women > 55
years of age
Ovaries detected clinically in 9% of women
> 200 lbs
Ovaries detected clinically in 12% of women with
a uterine weight > 200 grams
Ueland et al. Gynecol Oncol 99:400-403, 2005.
Screening

The identification of unrecognized disease
by the application of tests or examinations
to apparently well persons to distinguish
those who have a disease from those who
do not.
Long-term Results of a
Successful Screening Test
Decreased stage of detection
 Decreased case-specific mortality rate
 Decreased site-specific mortality rate

Characteristics of A Disease
Suitable for Screening
Serious consequences (morbidity/mortality)
 Effectively treated when diagnosed early
 High prevalence among screened
population
 Detectable preclinical phase

Disease Progression
Detection by Screening
Disease O
Preclinical Disease
Biologic
Onset
Symptom
Onset
Characteristics of Tests Suitable
for Screening
Safe
 Simple to perform/time efficient
 Cost-effective
 Acceptable to patients/non-invasive
 Valid : High Sensitivity
High Specificity
High Positive Predictive
High Negative Predictive Value

Statistical Definitions
Screening Test Result
Preclinical Disease
Present
Absent
A (TP)
B (FP)
Positive
D (TN)
C (FN)
Negative
Sensitivity = TP/(TP+FN) = A / A+C
PPV= TP/(TP+FP)= A / A+B
NPV= TN/(TN+FN)= D / C+D
Specificity = TN/(TN+FP) = D / B+D
Transvaginal Sonography (TVS)
Transvaginal Sonography (TVS)
Easy to perform
 Well-accepted
 Cost-effective when performed in screening
setting

The University of Kentucky
Ovarian Cancer Screening
Program
Initiated in 1987
 Eligible if  50 years old or  25 with
family history of ovarian cancer

TVS Screening Algorithm
TVS Screening
Normal
Abnormal
Repeat TVS
1 year
Repeat TVS
4-6 weeks
Criteria for Abnormal TVS
Volume > 20 cm3 premenopausal
 Volume > 10 cm3 postmenopausal
 Tumor complexity - any solid or papillary
projection into a cystic lumen

Pavlik, E.J., et. al. Relating Ovarian Size to Age, Menopausal Status, and Use of
Hormones. Gynecol. Oncol. 80:333-334, (2001).
TVS Screening Algorithm
Repeat Screen
Normal
Abnormal
Repeat TVS
1 year
Tumor morphology indexing,
CA-125, Color Doppler, Proteomics
Complex mass or rising MI
or elev. CA-125
Surgery
Repeat TVS
4-6 weeks
Cystic mass w/
normal CA-125
Repeat TVS 6 mos
Mi100.jpg
Morphology Indexing: MI = 0
 Cystadenoma
Morphology Indexing MI = 6
 Cystadenocarcinoma
Morphology Index (MI)
Preoperative MI was performed on 442
ovarian tumors (0-10)
 1/314 tumors with MI < 5 found to be
malignant
 52/127 tumors with MI  5 found to be
malignant
 Sens. 98.1%, Spec. 80%, PPV 40.1%,
NPV 99.7%

Ueland et al. Gynecol. Oncol. 91: 46-50 (2003).
The University of Kentucky
Ovarian Cancer Screening
Program 1987 - 2006
25,327 women screened
 120,569 free screens
 116,568 screening years
 364 patients (1.4%) with persisting ovarian
tumors operated upon

Histology of Ovarian Tumors
Discovered by TVS Screening (n=364)










Primary ovarian cancer
Serous cystadenoma
Endometrioma
Mucinous cystadenoma
Cystic teratoma
Fibroma/thecoma/Brenner
Leiomyoma
Hydrosalpinx/paratubal cyst
Other
Non-ovarian malignancies
44
153
30
19
18
25
4
25
39
7
Summary of Primary Ovarian
Cancers Detected by Screening
(N = 44)
Stage I – 28
 Stage II – 8
 Stage III – 8

The University of Kentucky
Ovarian Cancer Screening
Program 1987 - 2006
9 patients with ovarian cancer detected
within 12 months of negative screen (false
negative)
 NED = 6
 DOD = 3

The University of Kentucky
Ovarian Cancer Screening
Program 1987 - 2006
Statistical Data
Sensitivity = 85.0%
 Specificity = 98.8%
 Positive Predictive Value = 13.8%
 Negative Predictive Value = 99.9%

Survival of Ovarian Cancer
Patients in the Annually Screened
Group

2 yr – 92.1%

5 yr – 82.4%
Conclusions




TVS screening is safe, time-efficient, and wellaccepted by patients
Annual TVS screening causes a significant
decrease in stage at detection and case -specific
ovarian cancer mortality.
The cost of each screen is approximately $50
which is well within the range of other screening
methods
TVS screening is not effective in detecting
primary peritoneal cancer or ovarian cancer in
which the ovarian volume is normal
Unresolved Issues
Who should be screened?
 What is the optimal screening interval?
 What is the optimal screening algorithm?
